CLN Article

Managing Test Utilization with Physician Report Cards

In an environment of rising pressure to cut costs, many labs first targeted the most expensive send-out tests that can cost thousands of dollars. However, labs have found that the routine, low-cost tests add up, too. One organization that has tackled both ends of the spectrum is Group Health Cooperative (GHC), a large managed-care system in Washington State.

As both a health plan and a delivery system, GHC has tried a variety of programs to manage test utilization. For the more expensive genetic tests, individual decisions about when a test is appropriate are taken out of the lab's hands because GHC requires preauthorization from the health plan. But laboratorians still play an important role on a medical technology assessment committee, which includes a consumer, two epidemiologists, and physicians from across several disciplines, according to Kim Riddell, MD, a clinical pathologist and service line chief of clinical laboratories for GHC, who represents the lab on this committee (See Box, below). The committee's recommendation about the evidence on a new test goes to a medical directors' group that ultimately decides whether the test will be covered. To keep the list up-to-date, the committee revisits each decision every 2 years to consider new evidence. This system worked so well for GHC that the Washington State Department of Health adopted it to evaluate new technologies for coverage under its Medicaid program.

Group Health Cooperative Genetic Test Evaluation Criteria

  1. Regulatory approval: Has the test been cleared or approved by the FDA for the indication under review?
  2. Analytic validity: Is there sufficient scientific evidence to conclude that the test accurately identifies the particular genetic variant or mutation in questions, i.e., the test's sensitivity and specificity for the genetic variant?
  3. Clinical validity: Is there sufficient scientific evidence to conclude that the test accurately predicts a particular clinical outcome, i.e., is the presence of the variant or mutation associated with outcome?
  4. Risks of diagnostic test: Is there sufficient scientific evidence to conclude that the test or procedure itself, rather than the activity that follows, is not harmful to patients?
  5. Clinical utility: Is the test reasonably expected to lead to more appropriate patient management than if the test were not available?

Putting the health plan in charge of preauthorizing certain tests makes sense, according to Riddell. "I run the laboratory, but I don't have in-depth knowledge about how to administer health-plan decisions," she said. "If you're going to be consistent and administer these kinds of decisions fairly, it has to be a process done through the health plan. The lab can give advice, but we're talking about deciding who gets a service and who doesn't. I think that's really a health-plan function." Most genetic tests require prior authorization before the lab can draw a sample. In the beginning, there was a learning curve because the lab got hit with many complaints from doctors and patients about delays. Over time, the health plan has improved the process of adjudicating test preauthorizations and complaints are subsiding, she added.

While preauthorization takes care of some of the newest and most expensive tests, the high-volume, low-cost side of the menu is equally important, according to Riddell. She found that overutilization of these tests, like complete blood counts or chemistry panels, had become just as onerous. Riddell made the discovery in 2003 when GHC benchmarked the lab against other local labs. "Our unit costs were very competitive," she recalled. "But our utilization, as tests per 1,000 patients, was way off compared to other benchmarked services."

With a mandate to make the lab a more competitive and efficient part of the organization, Riddell set out on an intensive utilization management campaign that targeted high-volume tests that had no clear clinical need. The worst offenders, it turned out, were some of the 200 family physicians practicing in 20 outpatient clinics under GHC who ordered large numbers of tests for routine wellness visits.

Riddell began with an in-depth study of more than 200 patient charts and found that almost one third of testing came from these periodic health assessments, or wellness visits. "Basically, healthy patients were getting lab tests just for showing up," Riddell said. Only a few tests are actually recommended for wellness visits, she noted. For children, newborn screen, spun hematocrit, urinalysis, and urine culture; and for adults, cholesterol or lipid panels, Chlamydia¸ PAP smear, occult blood, and prostate specific antigen under shared decision-making between the physician and patient.

Riddell then focused her efforts on educating physicians, an initiative that continues to be worthwhile. The program consists of a newsletter to physicians that explains evidence-based guidelines behind wellness testing—for example, why tests like complete blood counts and thyroid stimulating hormone are not recommended. Riddell also sends each clinic a report card that shows how many of these tests each physician is ordering (See Box, below). Since initiating the program in 2003, Riddell has witnessed dramatic decreases for many of these tests. "I've found that family practice physicians generally buy into this," she said. "They appreciate being coached on good utilization." This year, the GHC quality leadership group wants to do more in this area.

Example of a Lab Utilization Report Card
Example of a Lab Utilization Report Card
*Data are de-identified

The next step will be taking a hard line on some of these tests and limiting coverage through the health plan, Riddell said. However, health plans must carefully navigate state and federal regulations when it comes to limiting coverage. "We'd like to be able to tell the physicians that if they order one of these unnecessary tests on a healthy person, it's not going to be covered and Group Health will not pay for it," she said. "Ideally, we want good align incentives between our health plan and good, basic test ordering practice."

Riddell noted that, currently, both private and public payers often incentivize overutilization and can seem to work against her efforts. "I'm hoping that payers take away the perverse incentives to order more. For example, the comprehensive metabolic panel with 14 tests is cheaper than the individual tests," Riddell commented. "So if we encourage doctors to order only individual tests when they need to, the patient ends up paying more out of pocket. In the eye of the patient, this looks bad and defeats the whole idea of wise ordering. We need to show patients that good ordering practice benefits them as well."